LIMIT OF DETECTION (LoD)

III. QUALITY ASSURANCE AND MANAGEMENT OF LABORATORY ACTIVITY IN THE FIELD
OF ENVIRONMENTAL MONITORING AND CONTROL INCLUDING SONIC AND ELECTROMAGNETIC
POLLUTION
Laboratory Exercises
III. 1. VALIDATION OF TEST METHODS
III.1.1. Theoretical Aspects
Validation means "confirmation by examination and prediction of
objective evidence that the particular requirements for a specified
intended use are fulfilled" (according to ISO 8402:1994).
Method validation means:
-The process of establishing the performance characteristics and
limitations of a method and the identification of the influences, which may
change these characteristics, and to what extent.
. Which analytes can be determined, in which matrices, in the
presence of which interferences?
. Within these conditions what levels of precision and accuracy can
be achieved?
-The process of verifying that a method is fit for a purpose, i.e. for
solving a particular analytical problem.
Verification means "confirmation by examination and prediction of
objective evidence proving that the specified requirements have been
fulfilled" (according ISO 8402:1994).
It is necessary to make the difference between validation and
verification.
Verification is applied for standardized methods and validation must be
made for:
. non-standard methods;
. laboratory - designed / developed methods;
. standard methods used outside their intended purpose;
. standard methods.
Validation studies for analytical methods typically determine the
following parameters:
. detection limit;
. quantification limit;
. working range;
. selectivity;
. sensitivity;
. robustness;
. recovery;
. accuracy;
. precision ;
. repeatability ;
. reproducibility.
The performance parameters being tested are selected depending on the
analytical requirements and based on the specifications from Table III.1.1. Table III.1.1. Analytical requirements and the corresponding
performance parameters. |Analytical requirements |Related performance |
| |parameters |
|- Qualitative or quantitative|Confirmation of identity, |
|answer? |selectivity/specificity, |
| |Limit of detection |
| |Limit of quantification |
|For the analyte present in |Recovery |
|more than one form, is | |
|important the extractable, | |
|free or total analyte? | |
|analyte(s) of interest and |Limit of detection |
|the most probable level (%, |Limit of quantification |
|?g g-1, ng g-1 etc.)? |Working range |
|Level of precision and |Recovery |
|accuracy, allowed uncertainty|Accuracy |
|degree. |Repeatability |
| |Reproducibility |
|Possible interferences |Selectivity/specificity |
|Comparison of results with |robustness |
|results from other |Reproducibility |
|laboratories? | |
|Comparison of the results |Accuracy |
|with external specifications?|Reproducibility | Limit of Detection (LoD) means:
- the lowest content that can be measured with reasonable statistical
certainty;
- the lowest concentration of analyte in a sample that can be detected,
but not necessarily quantified under the stated conditions of the test;
- the lowest analyte content, if actually present, that can be detected
and can be identified.
Where measurements are made for low concentrations of analyte (trace
analysis) it is important to know what is the lowest concentration of
analyte that can be confidently detected by the method. This problem must
be analyzed statistically and a domain of decision criteria must be
proposed.
It is normally sufficient to provide an indication of the level at which
detection becomes problematic.
For quantitative measurements, 10 independent blank samples (a) or 10
independent blank samples fortified at lowest acceptable concentration (b)
are analyzed, measured a single time each, and the mean value and standard
deviation (s) of the blank sample is calculated for each set of
measurements.
LoD is expressed as the analyte concentration corresponding to:
a) mean value of the blank sample + 3 s;
b) 0 + 3 s or the mean value of the blank sample + 4.65 s
For qualitative measurements, is sufficient a critical concentration
below which the specificity can not be identified. Thus, for a series of
concentration levels are analyzed blank samples injected with analyte. For
each concentration level it is necessary to make 10 independent repeated
measurements and a response curve of % positive or negative results versus
concentration should be constructed. From this curve can be established, by
interpolation, the threshold concentration at which the test becomes
unreliable.
Generally, the LoD, expressed in terms of concentration cL, or the
quantity qL, is derived from the smallest measurement xL, that can be
detected with reasonable certainty for a given analytical procedure. The
value of xL is calculated with the formula: xL = xbl + k sbl where : xbl is the mean value of the measurements for the blank sample of
reagents; sbl is the standard deviation of the measurements for the blank
sample of reagents ; k is a numerical factor chosen according to the
desired confidence level. Table III.1.2. Limit of detection
|LIMIT OF DETECTION (LoD) |
|Measurements |10 independent blank samples one time |
| |measured or |
| |or 10 blank samples fortified at |
| |lowest acceptable concentration , one |
| |time measured |
|Determination/Estimat|LoD = 3s + X |
|ion |in which: |
| |s = standard deviation for the blank |
| |or blank fortified with an analyte |
| |samples |
| |X = measured value or mean measured |
| |value |
|Optimum value |function of tested method type | Limit of Quantification (LoQ), known as Quantifiable Limit means:
- the content equal to or greater than the lowest concentration point on
the calibration curve;
- the lowest concentration of analyte in a sample that can be determined
with acceptable repeatability and accuracy;
- performance characteristics that mark the ability of a chemical
measurement process to adequately quantify an analyte.
The ability to quantify is generally expressed in terms of the signal
or analyte value that will produce estimates having a specific relative
standard deviation (RSD), commonly 10%. The formula of calculation is: LoQ = kQ?Q where: - ?Q is the standard deviation at that point; kQ is the multiple
whose reciprocal equals the RSD. The IUPAC recommended value for kQ is 10.
The following analyses will be made:
- 10 independent blank samples measured once each and the standard
deviation (s) is calculated. LoQ is expressed as the concentration of
the analyte corresponding to the a blank sample value + 10s;
- fortified aliquots of a blank sample at various analyte concentrations
close to the LoD and the standard deviation (s) of each concentration is
calculated. (s) is represented graphically against concentration and a
value to the LoQ is established by interpolation. Table III.1.3. Limit of Quantification |LIMIT OF QUANTIFICATION (LoQ) |
|Measurements |10 independent blanks one time measured|
| | |
| |or |
| |10 blank samples fortified at lowest |
| |acceptable concentration , one time |
| |measured |
|Determination |LoQ = 10s + X |
|/Estimation |where: |
| |s = standard deviation for the blank or|
| |blank fortified with an analyte samples|
| | |
| |X = measured value |
|Optimum value |function of tested method type | Working Range - the analyte concentration interval or the value for
which the method can be applied is determined. Within the working interval
can exist a linear response interval. Sometimes also a nonlinear response
range may be used , in case of a stable situation and calculation by
computer. Generally, linearity studying involves at least 10 different
concentrations / property values. Anywhere, in the working range, multi -
point (preferably 6+) calibration points will be necessary. It is important
to retain that the working range and linearity may be different for
different matrices due to the of interferences if they are not eliminated.
Table III.1.4. Working range. |WORKING RANGE